R help - Dec 2009 - (Grouped + Stacked) Barplot

Hi All,

I have googled and tried finding if someone has ever tried producing
(Grouped + Stacked) Barplot. I couldn't find one.

My data needs to be reshaped, but once it is done it would be something like
this:

Factor1    Factor2    Factor3    Value
A    X    P    10
A    X    Q    20
A    Y    P    20
A    Y    Q    5
A    Z    P    20
A    Z    Q    10
B    X    P    20
B    X    Q    10
B    Y    P    30
B    Y    Q    50
B    Z    P    10
B    Z    Q    20

There are three categorical variables (Factor1, Factor2, and Factor3 having
2,3, and 2 levels resp.). I'm trying to plot "Factor1" on x axis
-- each
level of this factor should have three (grouped) bars (one for each level of
Factor2). Which will further be stacked by each level of Factor3.

Can anyone guide me please, thanks,
~Xin

	[[alternative HTML version deleted]]

On Dec 2, 2009, at 10:25 PM, Xin Ge wrote:
> Hi All,
>
> I have googled and tried finding if someone has ever tried producing
> (Grouped + Stacked) Barplot. I couldn't find one.
I think you need to define some dummy Factor1 categories with zero  
heights and blank labels to separate your groups. Once that is done,  
you can just use typical stacked bar graphs.

(Another barplot function to consider:

http://pbil.univ-lyon1.fr/library/gregmisc/html/barplot2.html
  )
>
> My data needs to be reshaped, but once it is done it would be  
> something like
> this:
>
> Factor1    Factor2    Factor3    Value
> A    X    P    10
> A    X    Q    20
> A    Y    P    20
> A    Y    Q    5
> A    Z    P    20
> A    Z    Q    10
> B    X    P    20
> B    X    Q    10
> B    Y    P    30
> B    Y    Q    50
> B    Z    P    10
> B    Z    Q    20
>
> There are three categorical variables (Factor1, Factor2, and Factor3  
> having
> 2,3, and 2 levels resp.). I'm trying to plot "Factor1" on x
axis --
> each
> level of this factor should have three (grouped) bars (one for each  
> level of
> Factor2). Which will further be stacked by each level of Factor3.
>
> Can anyone guide me please, thanks,
> ~Xin

David Winsemius, MD
Heritage Laboratories
West Hartford, CT

A bar graph including both stacked and grouped bars will put lots of pretty
colors on the page and probably be eyecatching, but is unlikely to be the most
effective way to convey the actual meaning of the data.  I would recommend that
you explore other possibilities for doing the plot.  A dot plot may give you
what you want, see the dotchart2 function in the Hmisc package or the dotplot
function in the lattice package.

-- 
Gregory (Greg) L. Snow Ph.D.
Statistical Data Center
Intermountain Healthcare
greg.snow at imail.org
801.408.8111

> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-
> project.org] On Behalf Of Xin Ge
> Sent: Wednesday, December 02, 2009 8:25 PM
> To: r-help at r-project.org
> Subject: [R] (Grouped + Stacked) Barplot
> 
> Hi All,
> 
> I have googled and tried finding if someone has ever tried producing
> (Grouped + Stacked) Barplot. I couldn't find one.
> 
> My data needs to be reshaped, but once it is done it would be something
> like
> this:
> 
> Factor1    Factor2    Factor3    Value
> A    X    P    10
> A    X    Q    20
> A    Y    P    20
> A    Y    Q    5
> A    Z    P    20
> A    Z    Q    10
> B    X    P    20
> B    X    Q    10
> B    Y    P    30
> B    Y    Q    50
> B    Z    P    10
> B    Z    Q    20
> 
> There are three categorical variables (Factor1, Factor2, and Factor3
> having
> 2,3, and 2 levels resp.). I'm trying to plot "Factor1" on x
axis --
> each
> level of this factor should have three (grouped) bars (one for each
> level of
> Factor2). Which will further be stacked by each level of Factor3.
> 
> Can anyone guide me please, thanks,
> ~Xin
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-
> guide.html
> and provide commented, minimal, self-contained, reproducible code.